INDIAN PEDIATRICS 45 VOLUME 51__JANUARY 15, 2014

RRRRR EEEEE SSSSS EEEEE AAAAA RRRRR CCCCC HHHHH B B B B B RRRRR IIIII EEEEE FFFFF

Modifiable Factors for Prevention of Childhood MortalityVIDUSHI MAHAJAN, AMARPREET KAUR, AMIT SHARMA, CHANDRIKA AZAD AND VISHAL GUGLANIFrom Department of Pediatrics, Government Medical College and Hospital, Sector 32, Chandigarh, India.

Objective: To know the disease-related causes of child mortality and identify sociallymodifiable factors affecting child mortality among hospitalized children aged >1 month-18years in a referral hospital of North India. Methods: Causes of death (ICD-10 based) wereextracted retrospectively from hospital files (n=487) from 17 March 2003 to 30 June 2012.Modifiable factors were prospectively studied in 107 consecutive deaths from 6 October2011 to 30 June 2012. Results: Pneumonia, CNS infections and diarrhea were the mostcommon disease-related causes of child mortality. Conclusions: Amongst modifiablefactors, administrative issues were most common followed by family-related reasons andmedical-personnel related problems.

Keywords: Audit, Child deaths, India, Prevention.

Correspondence to:Dr Vidushi Mahajan,Assistant Professor, Department ofPediatrics, Government Medical Collegeand Hospital, Sector 32, Chandigarh,India.vidushimahajan2003@yahoo.co.inReceived: February 1, 2013;Initial review: March 04, 2013;Accepted: July 05, 2013.

India hosts maximum (24%) number of deaths inunder-5 children occurring worldwide [1].Disease-related or ‘biological’ factors related tochild mortality are studied extensively. Majority of

childhood deaths in India are attributed to infections,particularly pneumonia and diarrhea [1- 6]. Certain non-biological causes (e.g. administrative, medical personnel,family-related factors) may also contribute towards childmortality [7]. We planned this study to evaluate disease-related causes and modifiable factors of child mortalityamong hospitalized children aged >1 month-18 years whodied in a tertiary care referral teaching hospital of NorthernIndia.

METHODS

The study had a mixed design; disease-related causes ofmortality were analyzed retrospectively (17 March, 2009 to30 June, 2012) and socially modifiable factors wereidentified prospectively (6 October, 2011 to 30 June, 2012) inchildren (age 1 mo-8 y), who died in Pediatric emergencyward (PEW) and Pediatric Intensive care unit (PICU). Ethicalapproval was obtained.

To study the disease-related causes of mortality, weextracted the relevant clinical details and final diagnosisfrom hospital files of the study population. We excludedany missing case records. The ‘primary cause’ of deathwas the probable cause that finally led to death of the child[8]. The causes of death were ICD-10 based.

To study the modifiable causes of child mortality, weenrolled all critically sick children admitted in PEW andPICU. A list of modifiable factors was developed a. priori,

which were defined as events, actions or omissionscontributing to death of a child and which, by means ofinterventions, could be modified [9]. These factors werecategorized as: (A) Family/caregiver-related problemswhich included - (i) delay in getting medical attention (e.g.lack of transport, girl child, delayed referral by primary carephysician, inability to recognize danger signs, maternal illhealth), (ii) treatment by quacks/faith healers, (B) Medicalpersonnel-related factors included - (i) clinical assessmentissues (delay in detection of signs, delayed referral bytreating team, alternative diagnosis not considered), (ii)monitoring issues, and (iii) case management (prescriptionerror, delay in institution of specific management) at ourhospital; and (C) Adminis-trative factors included - (i)shortage of staff (residents, nurses), (ii) shortage/non-functioning of equipment(s), (iii) lack of specializedlifesaving care e.g. dialysis, surgical procedure etc, (iv) lackof PICU beds/ ventilators, (v) communication gap betweenmedical staff, (vi) Lack of drugs, blood products, and (vii)lack of policy. Resident doctors recorded these factorsduring history-taking, which were cross-checked by aconsultant pediatrician. The staff was periodically primed torecord all study variables.

To identify modifiable factors audit meetings were heldfortnightly, using death-audit profoma and patient records.Each meeting was attended by at least three consultingpediatricians (one primary consultant who managed thecase and two unrelated consultants), concerned residentdoctors and nursing staff, where deemed necessary.Consensus on causes of death, contributing conditionsand modifiable factors were reached.

Published online: August 05, 2013. PII: S097475591300104

INDIAN PEDIATRICS 46 VOLUME 51__JANUARY 15, 2014

MAHAJAN, et al. MODIFIABLE FACTORS FOR CHILDHOOD MORTALITY

Proportion of disease-related causes of mortality; andproportion of modifiable factors related to child mortalitywere the two outcome variables. Descriptive statistics wasused to describe baseline demographic variables andmodifiable factors. Data were analysed by Excel and SPSS V.17.0.RESULTS

There were 5815 admissions (>1 month) during the studyperiod. Of these, 493 children died (case fatality rate 8.4%).We excluded six cases whose files could not be traced. Wetherefore analyzed 487 deaths [237 (48.6%) one-month to 1-year, 138 (28.3%) in 1-5 years and 112 (23%) in children >5years] for disease-related causes of mortality. Pneumonia,CNS infections and diarrhea were leading disease-relatedcauses of mortality (Fig. 1). Severe malnutrition (42%) wasthe major contributing cause [median z–score: -1.94 (IQR -3.37 to -0.68)].

We studied modifiable factors amongst 107 (5% males)consecutive deaths. Their median (IQR) age was 12 (5, 60)months and weight was 8 (5, 15) kg. 43% had shock atpresentation, as defined by AHA [10,11], and 6% had acardiac arrest either before or at presentation to PEW.Seventy one percent required mechanical venti-lation

within one hour of presentation. Median hospital stay was32 (IQR-10, 101) hours. Majority (64%) of the parents ofstudy children lived in villages, were illiterate (mothers-46%, fathers-29%) and worked as manual labourers/ dailywagers (61%).

Amongst modifiable factors, administrative issueswere most common (universal) followed by family/caregiver-related factors (72%) and medical personnel-related factors (41%). Shortage of medical personnelespecially senior residents and nursing staff remained aconstant feature throughout. Among medical personnel-related factors, improper monitoring was single mostprevalent factor. (Table I).

DISCUSSION

Pneumonia, CNS infections and diarrhea were main causesof disease-related mortality and administrative issuesfollowed by family-related reasons were most commonmodifiable factors in our study.

Our study results were in accordance with national andglobal estimates of child mortality [1-6] and national auditreports, evaluating modifiable factors, published fromSouth Africa [7-9]. We found a considerable proportion of

(a) (b)

FIG. 1 Causes of deaths in (a) children of all ages (N=487) (b) children 1-12 months (N=237) (c) children 1-5 years (N=138) (d) childrenabove 5 years (N=112).

(c) (d)

INDIAN PEDIATRICS 47 VOLUME 51__JANUARY 15, 2014

MAHAJAN, et al. MODIFIABLE FACTORS FOR CHILDHOOD MORTALITY

deaths due to CNS infections and tuberculosis in our study,which is expected because of the referral hospital setting.

Although administrative issues were presentuniversally, majority of them are related to theinfrastructure, availability of healthcare personnel andequipments. These factors, though modifiable, are relatedto health resource allocation and budget constraints.However, a few administrative factors e.g. availability ofdrugs, and unit policy decisions can be locally modified.There was a high incidence of monitoring issues which islinked to the poor doctor: patient (1:40-1:70)/nurse: patient(1:20-1:30) ratio, with bed occupancy >100% during studyperiod.

We acknowledge the limited sample size of our study

and mixed retrospective-prospective study design.However, our study population was both rural and urbanincluding slums, thereby giving an insight to deathsoccuring in all sections of society.

Family-related factors were present in more than two-third of child deaths. Largely, children who died were verysick at admission, which underscores the importance ofearly health seeking. Majority of our population were dailywagers with poor literacy levels, which could contribute todelayed illness recognition.

REFERENCES

1. The latest estimates on child mortality generated by the UNInter-agency Group on Child Mortality Estimation(IGME): Levels and Trends in Child Mortality, Report2012 (13 September 2012). http://www.childinfo.org/mortality.html. Accessed on 14 January, 2013.

2. Million Death Study Collaborators, Bassani DG, Kumar R,Awasthi S, Morris SK, Paul VK, Shet A, et al. Causes ofneonatal and child mortality in India: a nationallyrepresentative mortality survey. Lancet. 2010;376:1853-60.

3. Black RE, Cousens S, Johnson HL, Lawn JE, Rudan I,Bassani DG, et al. Global, regional, and national causes ofchild mortality in 2008: a systematic analysis. Lancet.2010;375:1969-87.

4. Black RE, Morris SS, Bryce J. Where and why are 10million children dying every year? Lancet. 2003;361:2226-34.

5. Jones G, Steketee RW, Black RE, Bhutta ZA, Morris SS;Bellagio Child Survival Study Group. How many childdeaths can we prevent this year? Lancet. 2003;362: 65-71.

6. Million Death Study Collaborators, Morris SK, BassaniDG, Awasthi S, Kumar R, Shet A, Suraweera W, et al.Diarrhea, pneumonia, and infectious disease mortality inchildren aged 5 to 14 years in India. PLoS One. 2011;6:e20119.

7. Krug A, Patrick M, Pattinson RC, Stephen C. Childhooddeath auditing to improve paediatric care. Acta Paediatr.2006;95:1467-73.

8. Krug A, Pattinson RC, Power DJ. Saving children - an auditsystem to assess under-5 health care. S Afr Med J.2004;94:198-202.

9. Pattinson, RC. Practical application of data obtained from aPerinatal Problem Identification Programme. S Afr Med J.1995;85:131-2.

10. Kleinman ME, Chameides L, Schexnayder SM, Samson RA,Hazinski MF, Atkins DL, et al. Part 14: Pediatric AdvancedLife Support: 2010 American Heart Association Guidelinesfor Cardiopulmonary Resuscitation and EmergencyCardiovascular Care. Circulation. 2010:122: S876-908.

11. Kleinman ME, de Caen AR, Chameides L, Atkins DL, BergRA, Berg MD, et al. Part 10: Pediatric Basic and AdvancedLife Support: 2010 International Consensus onCardiopulmonary Resuscitation and Emergency Cardio-vascular Care Science with Treatment Recommendations.Circulation. 2010;122:S466-515.

TABLE I MODIFIABLE FACTORS AMONG 107 DEATHS

Modifiable Factors* N (%)

Family caregiver relatedTransport problem 15 (14)Female child 5 (5)Delayed referral 34 (32)Delay in illness recognition 45 (42)Maternal Ill-health 4 (4)Quacks and faith healers 11 (10)No family related issues 30 (28)

Medical personnel relatedAssessment

Delayed detection 5 (5)Alternative diagnosis not considered 2 (2)Delay in specific management 11 (11)Case monitoring 20 (19)

Case managementDelayed referral 1(5)Prescription error

No medical personnel issues 63 (59)Administrative factors

Bed or ventilator unavailability 57 (53)Lack of specialized care 3 (3)Lack of equipment 12 (11)Lack of medical personnelCommunication problems 1Lack of drugs and blood products etc 2Lack of policy 1

*More than one modifiable factor were present in some studysubjects; #Lack of medical personnel was a constant featurethroughout the study.